Nutrition

Diet pattern matters more than any individual macro, micro, or "superfood." The most consistently validated dietary patterns for longevity — Mediterranean, MIND, and DASH (Dietary Approaches to Stop Hypertension) — share a common core: lots of plants, fish, nuts, legumes, olive oil; minimal ultra-processed food, sugar-sweetened beverages, and processed and red meat.

Nuance matters too: protein adequacy in midlife protects muscle and bone, which conflicts with the popular "low protein extends lifespan" reading of animal data — the resolution lives under Protein. Fasting improves blood-sugar control but does not beat continuous calorie restriction for weight loss; on lifespan, no human trial has ever tested it either way.

What the evidence actually supports

Strong:

• Mediterranean diet pattern is the best-evidenced way to eat for a long life. Pooled cohort data are consistent: a 2024 review of 28 studies and nearly 680,000 adults found lower all-cause and cardiovascular mortality with closer adherence.^[1] The landmark trial, PREDIMED, randomised 7,447 Spanish adults at high cardiovascular risk and found roughly 30% fewer heart attacks and strokes.^[2] Read that trial with its history attached: the original 2013 paper was retracted and republished in 2018 after an audit found randomisation irregularities affecting about a fifth of participants — some households and clinics were assigned as blocks rather than as individuals — and the olive-oil and nut industries part-funded the work. The 30% survived re-analysis, which is why the pattern still leads this list, but Dietary patterns rates the trial itself Moderate rather than Strong, and sets Mediterranean, MIND, DASH, and the Blue Zones side by side.
• MIND diet — the Mediterranean–DASH Intervention for Neurodegenerative Delay, a hybrid of the two emphasising leafy greens and berries — in the original 923-person cohort, the people sticking most closely to it had about 53% lower Alzheimer's risk; the middle group, about 35% lower.^[3] A subsequent randomised trial in 604 older adults showed no advantage over a mildly calorie-restricted control diet, leaving the observational signal stronger than the trial one.^[4] See Dietary patterns.
• Sugar-sweetened beverages are the single worst delivery vehicle for added sugar. Each daily 355 mL serving tracks with roughly 7% higher all-cause and 10% higher cardiovascular mortality in pooled cohorts,^[5] and added sugar in general speeds up validated biological-age clocks in a dose-response pattern.^[6] A wrinkle worth knowing: in the Framingham cohort it was diet soft drinks, not sugared ones, that carried the striking stroke and dementia signal.^[7] See Sweeteners for what added sugar, fructose, and high-fructose corn syrup actually do, and Sugar substitutes for the non-caloric options — artificial sweeteners and the World Health Organization (WHO) and cancer-agency verdicts on them, the erythritol and xylitol cardiovascular signal, and where stevia, monk fruit, and allulose land.
• Processed meat is classified by the WHO's cancer agency (the International Agency for Research on Cancer, IARC) as a known human carcinogen — Group 1, the same category as tobacco and asbestos. That category describes how confident the agency is that the substance causes cancer, not how much cancer it causes, and the two are wildly different here. Each 50 g a day — about two rashers of bacon — raises colorectal-cancer risk by 16–18%, which sounds alarming until it is put in absolute terms: against a lifetime baseline risk near 5%, that moves an individual to roughly 6%. Worldwide, high-processed-meat diets are estimated to account for about 34,000 cancer deaths a year, against roughly a million for tobacco.^[8] Unprocessed red meat sits one tier down as a probable carcinogen (Group 2A), with about 15% higher colorectal-cancer risk comparing the highest intakes with the lowest — against the same 5% baseline, a shift to roughly 5.7%.^[9] It also carries roughly 11% higher cardiovascular risk per 100 g a day.^[10] Parallel signals run through type 2 diabetes^[11] and dementia.^[12] The mechanisms are unusually well worked out — a sugar molecule in beef, pork, and lamb that humans lost the ability to make and now mount a low-grade immune response against; a gut-bacterial metabolite of meat that promotes arterial plaque; and iron-catalysed oxidation in the colon — and they, plus the cooking methods that cut carcinogen formation, the 2019 NutriRECS guideline panel that told people to carry on eating red meat, and the paradox around APOE4 (the main genetic risk variant for Alzheimer's), are under Red and processed meat. The wider list of harm signals — sugar, processed and red meat, alcohol — and the things that aren't as harmful as commonly claimed live under Foods to limit.
• Fat quality outweighs fat quantity. What replaces a fat decides the outcome: swapping saturated fat for polyunsaturated fat cuts cardiovascular disease by roughly 30%, while swapping it for refined carbohydrate achieves nothing.^[13] Trials that cut saturated fat and specify the replacement reduce cardiovascular events; trials that just say "eat less fat" are null.^[14] In practice: eliminate industrial trans fats (largely already done in regulated markets), anchor cooking on extra-virgin olive oil, and get omega-3s from fatty fish twice a week or about 1 g a day of combined EPA and DHA, the two long-chain omega-3 fats in fish oil — a dose at which a post-hoc analysis of the DO-HEALTH trial found modestly slower biological-age clocks, in a trial that missed all six of its primary clinical endpoints.^[15] And don't fall for the seed-oil panic: at ordinary home-cooking intakes the omega-6 case against them is disconfirmed, and what harm exists belongs to repeatedly reheated commercial frying oil and to the ultra-processed foods these oils usually arrive inside. The full picture, including eggs and the omega-6 to omega-3 ratio, is under Dietary fats.
• Adequate dietary protein in older adults preserves muscle, reduces age-related muscle loss (sarcopenia), and improves recovery from illness or surgery. The international expert consensus (PROT-AGE) sets a floor of 1.0–1.2 g of protein per kg of body weight per day for healthy older adults, rising to 2.0 g/kg during acute recovery.^[16] A floor is not a target — Protein argues for 1.0–1.5 g/kg past 65, and explains why blanket "low protein extends life" advice misreads the midlife and older-adult evidence.
• Habitual fermented dairy (yogurt, kefir, traditionally aged cheese) tracks with lower all-cause, cardiovascular, and cancer mortality in pooled cohorts — the most consistent epidemiologic signal in the fermented-food literature, though the effect is modest at roughly 6–7%, and for yogurt specifically the cardiovascular association did not reach statistical significance.^[17] See Fermented foods for why the food matrix matters more than live culture counts.

Moderate:

• Ultra-processed food carries one of the largest harm signals in modern nutrition science, though the certainty is lower than the headline numbers suggest. A 2024 umbrella review pooling 45 meta-analyses across roughly 10 million people linked high intake to 32 adverse outcomes — but under GRADE, the standard system for rating how much confidence a body of evidence warrants, only the cardiovascular-mortality and type-2-diabetes associations reached the top tier; most were low or very low.^[18] The defensible estimate is about 15% higher all-cause mortality in the highest intake category versus the lowest — the often-quoted 62% is a single-cohort outlier — while the per-10%-of-calories increment is genuinely contested, running from about 3% to about 10% across the two 2025 dose-response meta-analyses.^[19][20] The strongest causal evidence is a 2025 randomised crossover trial in which both diets met UK dietary guidelines and were matched on fat, saturated fat, protein, carbohydrate, salt, fibre, and produce; the minimally processed arm produced about twice the weight loss.^[21] It is 55 people over eight weeks and has drawn published critiques over energy density and author conflicts, so it points the way rather than settling it. Residual confounding from healthy-user bias and reverse causation remains a live alternative explanation for the cohort data. See Ultra-processed food.
• Time-restricted eating with an early eating window ranks ahead of late-window eating for fasting insulin and body composition at matched calories — though that ranking rests on indirect comparison across trials, and the best head-to-head crossover since found no difference.^{[22] [23]} The full intermittent-fasting, alternate-day, and fasting-mimicking landscape is covered under Fasting.
• Fasting and continuous calorie restriction produce similar weight loss and metabolic improvements over the long run. A 2025 BMJ pooled analysis of 99 randomised trials in 6,582 participants found alternate-day fasting had only a tiny edge of about 1.3 kg in short trials, which vanished entirely beyond 24 weeks;^[24] a 2026 Cochrane review of 22 trials in 1,995 adults reached the same conclusion.^[25] On lifespan itself there is nothing to compare: no human trial has ever measured it.
• High polyphenol intake (berries, olive oil, dark chocolate, tea, herbs) — a 2024 review pooling 7 cohort studies across nearly 180,000 adults found roughly 7% lower all-cause mortality at higher intake,^[26] and an observational re-analysis of the PREDIMED trial flagged above found 37% lower mortality in the highest-polyphenol fifth of participants versus the lowest — a comparison between self-selected groups, not between randomised arms.^[27]
• Fiber is the best-evidenced marker of carbohydrate quality — graded moderate certainty, against low-to-very-low for glycemic index. People eating the most fiber have roughly 15–30% lower all-cause mortality, coronary heart disease, stroke, type 2 diabetes, and colorectal cancer, with a dose-response that keeps falling through 25–29 g a day and beyond.^[28] The honest caveat is large: this is all observational, and no randomised trial has ever shown that adding fiber prevents anything — the supplement trials are null, and one increased the return of precancerous colon polyps. Eat legumes, whole grains, vegetables, fruit, nuts and seeds; don't buy it in a tub. See Fiber.
• Blood-sugar control after meals is a well-evidenced lever in diabetes management; in people without diabetes the long-term outcome benefit of flattening glucose spikes is unproven. Individuals respond very differently to the same standardised foods: the 800-person Zeevi cohort predicted personal glucose responses from gut microbiome, blood parameters, and habits,^[29] and a smaller 57-person Stanford study coined "glucotype" for a person's characteristic pattern of glucose swings.^[30] Eating vegetables and protein before starches ("carbs last"), choosing sourdough, adding vinegar, and eating earlier in the day flatten the curve without cutting carbohydrates — see Glycemic index.
• GLP-1 receptor agonists — glucagon-like peptide-1 drugs including semaglutide (Ozempic, Wegovy) and tirzepatide (Mounjaro, Zepbound) — produce durable double-digit weight loss and, in adults with overweight or obesity and established cardiovascular disease, a 20% reduction in heart attacks, strokes, and cardiovascular deaths — 6.5% of that group had such an event over about three years, against 8.0% on placebo — a benefit that appears largely independent of how much weight was actually lost.^[31] They are also the first drug class to move validated biological-age clocks in a randomised trial — but that result is a post-hoc analysis of 84 people in a 32-week trial not designed to test it, in a population selected for accelerated aging, with the clocks measured by the vendor that sells them, and it has not yet been peer-reviewed. The catch that matters most day to day: substantial muscle and bone loss without aggressive protein intake and resistance training. See GLP-1 receptor agonists.

Weak / preliminary:

• "Optimal" macronutrient ratios — wide ranges are compatible with longevity if the food quality is high. Plant-forward eating splits into two patterns: a healthful version (whole grains, legumes, fruit, vegetables, nuts) and an unhealthful one (refined grains, fruit juices, sweets). Higher adherence to the healthful pattern tracks with lower mortality; higher adherence to the unhealthful pattern tracks with higher mortality — even though both technically count as "plant-based."^[32] "Plant-based" alone is not the lever; food quality is.
• Many specific superfoods (turmeric, açaí, and the rest) have animal or mechanistic data without robust human trials.
• The capacity to switch fuels between glucose and fat — metabolic flexibility — is a plausible aging-resilience marker, trained by the things on this page rather than by anything specific to it.

Caution:

• Erythritol and xylitol, the dominant sugar alcohols in "keto" and "sugar-free" products, carry a real cardiovascular signal at ordinary intakes — enough that anyone with existing cardiovascular disease should avoid them routinely. See Sugar substitutes.
• Over-the-counter continuous glucose monitors now let healthy people watch their own blood sugar directly. The device measures a lagged, brand-dependent signal, most of the "spikes" it flags are normal physiology, and no trial shows that reacting to them improves health in people without diabetes — while the anxiety and over-restriction failure mode is real. A two-week experiment, not a lifelong wearable.

Topics covered in depth

What to eat

• Dietary patterns — Mediterranean, MIND, DASH, Nordic, and the Blue Zones, side by side, with what each one is actually built on.
• Protein — how much, from what, at what age, and why the "low protein extends life" argument does not survive contact with the midlife evidence.
• Dietary fats — the substitution effect, trans fats, olive oil, eggs, and why the seed-oil panic collapsed.
• Fiber — the strongest observational signal in nutrition, sitting beside a randomised record that is empty and, once, negative.
• Fermented foods — yogurt, kefir, kimchi, natto; why the food matrix beats live culture counts.
• A sample longevity week — seven days of meals that hit every target on this page at once, with older-adult, GLP-1, training, and vegetarian variants.

What to limit

• Ultra-processed food — the NOVA classification, its critics, and the trials that tried to isolate processing from nutrition.
• Red and processed meat — the processed/unprocessed split, its mechanisms, and how cooking method changes the exposure.
• Foods to limit — the full harm list, plus the things that turn out not to belong on it.
• Sweeteners — added sugar, fructose, and high-fructose corn syrup.
• Sugar substitutes — artificial sweeteners, sugar alcohols, stevia, monk fruit, allulose, and the erythritol signal.

When and how you eat

• Fasting — time-restricted eating, alternate-day fasting, and fasting-mimicking diets, measured against what the trials show.
• Glycemic index — food order, vinegar, cook-and-cool starches, sourdough, and why the index itself misleads.
• Metabolic flexibility — the capacity to switch between burning glucose and fat, and how it is trained.
• Continuous glucose monitors — what an over-the-counter sensor can and cannot tell a healthy adult.

Practical nutrition principles (evidence-weighted)

1. Default to a Mediterranean pattern. Vegetables, legumes, fish, nuts, olive oil, whole grains, fruit. Optional moderate dairy (yogurt, cheese), modest poultry. This is the single most-evidenced dietary intervention; see Dietary patterns.
2. Eat fish 2+ times/week (especially fatty fish: salmon, sardines, mackerel) or take EPA and DHA if not — see Omega-3.
3. Hit protein targets. Roughly 1.2–1.6 g of protein per kg of body weight per day for active midlife adults, and 1.0–1.5 g/kg past 65 — that range starts lower only because it covers all older adults rather than just active ones; if you train, sit in its upper half. Per meal, aim for about 0.4 g/kg (~25–30 g, rising to 30–40 g past 65 because older muscle responds less to a given dose), spread across 3–4 meals. See Protein.
4. Limit added sugar to under 10% of daily calories (World Health Organization) or about 25 g per day for women and 36 g for men (American Heart Association). The 2025–2030 US Dietary Guidelines take a different tack — no daily percentage ceiling, but no more than 10 g of added sugar in any single meal.^[33] The American Heart Association numbers are the stricter of the two and the ones worth aiming at — cap the daily total and avoid taking it all in one hit. Read Sweeteners for what added sugar does, and Sugar substitutes for the diet sweeteners and the erythritol and xylitol cardiovascular signal.
5. Limit ultra-processed food. The highest intakes carry about 15% higher all-cause mortality than the lowest; each 10% of daily calories from ultra-processed food adds somewhere between about 3% and 10%, which is as precise as the evidence currently allows. A 2025 randomised trial suggests — though does not prove — that some harm persists even when both diets meet dietary guidelines and are matched on fat, salt, and fibre. See Ultra-processed food.
6. Limit red meat to 1–3 servings/week; minimize processed meat. Replace with fish, legumes, poultry, nuts, or eggs — not refined carbohydrates, which cancels the benefit. See Foods to limit.
7. Eat 25–30 g of fiber a day, from food rather than supplements. Getting from the typical 16 g to 25 matters far more than getting from 28 to 30. Strongest single nutrient signal for cardiovascular and colorectal cancer prevention, though the evidence behind it is observational.
8. Front-load eating. Larger breakfast and lunch, lighter dinner, last meal by 18:00–19:00 where the schedule allows and at minimum 2–3 hours before bed. Aligns with circadian biology and produces measurably better fasting insulin than late-window eating at matched calories.
9. Don't smoke (smoking and nicotine); drink minimally (alcohol); hydrate adequately (water).

→ See a sample longevity week for a concrete 7-day menu that hits the protein, fibre, fish, legume, fermented-food, and meal-sequencing targets simultaneously, with notes for older-adult, GLP-1, training, and vegetarian/vegan variants.

What's overhyped

• Carnivore, ketogenic, and zero-carb diets for general longevity — short-term metabolic improvements in some populations, but long-term outcome data are absent or unfavorable. The most striking recent signal comes from mice bred to be deficient in DNA repair, an accelerated-aging model: near-doubling protein while cutting carbohydrate shortened median lifespan by 18% in males and 36% in females, with liver tissue showing inflammation and injury signatures.^[34] That is a sensitised model rather than a normal one, and it cuts both ways — severe protein restriction extended lifespan in the same mice, which is not advice this site would give a human either.
• "Detoxes" and cleanses — no clinical evidence for benefit; the liver and kidneys handle this.
• Specific superfoods marketed for individual health claims — diversity beats specialization.
• Most "anti-inflammatory diet" supplement protocols — anti-inflammatory eating works through the pattern, not concentrated extracts.

---

For the full ordered action list drawn from every pillar, see the healthspan long list.